Opportunistic beamforming with limited feedback: multi-user mode switching analysis

© 2014 Jiatian LiangThe explosive growth in data traffic in recent years has made the Multi-User Multiple-Input Multiple-Output (MU-MIMO) technique an important feature of the physical layer in modern wireless communication systems. MU-MIMO technique enables a base station (BS) with multiple antennas to communicate with each user simultaneously in order to optimise the sum rate. The technique requires accurate Channel State Information (CSI) at the user end. It is not practical to feedback a large amount of data in the uplink channel. Thus, it is important to focus on limited feedback schemes. In opportunistic beamforming, the sum rate is limited even as the transmission power is increased linearly in MU-MIMO. In contrast, the sum rate increases linearly with transmission power in single-user MIMO. In order to achieve the optimal sum rate, a method of mode switching is studied in this thesis. As limited feedback is considered, an optimal quantiser is essential for the accurate feedback of CSI. The method of developing an optimal quantizer is examined. The proposed optimal quantization scheme significantly outperformed the uniform quantization scheme. User scheduling has always been a challenging problem when the sum rate and fairness of scheduling are considered. This thesis investigates the impact of distance between transmitters and receivers, which is modelled by path loss. The study shows that users at the cell edge have little chance of being scheduled. A novel approach is presented to achieve a given fairness requirement while maintaining optimal capacity within the fairness constraints. Additionally, a multiple beamforming vectors scheme is presented to improve the sum rate when there are only a small number of users in the cell. The proposed approach greatly improves throughput when the number of users is small, regardless of transmission power. This thesis introduces a number of novel ideas and approaches to enhance the performance of a multi-user MIMO system. The derived expressions illustrate basic principles and can be utilised in more complex and practical scenarios